Artificial intelligence (AI) and machinelearning (ML) have become ubiquitous tools in the modern era. As more technologies become dependent on AI and ML, there is a greater push to understand how to use Explainable A...
详细信息
ISBN:
(纸本)9783031358906;9783031358913
Artificial intelligence (AI) and machinelearning (ML) have become ubiquitous tools in the modern era. As more technologies become dependent on AI and ML, there is a greater push to understand how to use Explainable AI (XAI) to help human users understand the underlying decision processes of the technology. We report on a laboratory-controlled experiment focused on trust and understanding comparing different types of XAI explanations with a recommendation system while controlling for participant gender and educational background. We found statistically significant interactions of both gender and educational background. As a result, we conclude that there is not one particular way to show explanations that is best for all audiences because both gender and the person's education and professional background play a part in determining their trust and understanding in the system. In addition, although a number of publications have promoted visual word clouds as a way to show XAI, our participants rejected that type of explanation and preferred textual explanations.
the proceedings contain 12 papers. the special focus in this conference is on Intelligent Edge Processing in the IoT Era, and Smart Governance for Sustainable Smart Cities. the topics include: Continuous Measurement o...
ISBN:
(纸本)9783031359811
the proceedings contain 12 papers. the special focus in this conference is on Intelligent Edge Processing in the IoT Era, and Smart Governance for Sustainable Smart Cities. the topics include: Continuous Measurement of Air Pollutant Concentrations in a Roadway Tunnel in Southern Italy;rating Urban Transport Services Quality Using a Sentiment Analysis Approach;edge Computing with Low-Cost Cameras for Object Detection in Smart Farming;Evaluating Maximum Operating Distance in COTS RFID TAGS for Smart Manufacturing;philippine Stock Direction Forecasting Utilizing Technical, Fundamental, and News Sentiment Data;ioT Architectures for Indoor Radon Management: A Prospective Analysis;adversarial Training for Better Robustness;integrating computer Vision and Crowd Sourcing to Infer Drug Use on Streets: A Case Study with 311 Data in San Francisco;machinelearning Approach to Crisis Management Exercise Analysis: A Case Study in SURE Project;quantitative Evaluation of Saudi E-government Websites Using a Web Structure Mining Methodology.
Withthe development of big data technology, it is necessary to quickly and accurately mine the information contained in data. In this paper, we study the selection of kernel function types and the selection of kernel...
详细信息
Practical quantum computing (QC) is still in its in-fancy and problems considered are usually fairly small, especially in quantum machinelearning when compared to its classical counterpart. Image processing applicati...
详细信息
Artificial Intelligence, particularly in machinelearning and related research areas such as Operational Research, currently faces a reproducibility crisis. Researchers encounter difficulties reproducing key results d...
详细信息
the availability of learning tools in schools that are accessible to low-income households is generally very limited when compared to most private schools that cater to the well-off masses. Contrarily, the accessibili...
详细信息
Human integration in machinelearning can take place in various forms and stages. the current study examines the process of feature selection, with a specific focus on eliciting and aggregating feature rankings by hum...
详细信息
ISBN:
(纸本)9783031358906;9783031358913
Human integration in machinelearning can take place in various forms and stages. the current study examines the process of feature selection, with a specific focus on eliciting and aggregating feature rankings by human subjects. the elicitation is guided by the principles of expert judgment elicitation, a field of study that has investigated the aggregation of multiple opinions for the purpose of mitigating biases and enhancing accuracy. An online experiment was conducted with 234 participants to evaluate the impact of different elicitation and aggregation methods, namely behavioral aggregation, mathematical aggregation, and the Delphi method, compared to individual expert opinions, on feature ranking accuracy. the results indicate that the aggregation method significantly affects the rankings, with behavioral aggregation having a more significant impact than mean and median aggregation. On the other hand, the Delphi method had minimal impact on the rankings compared to individual rankings.
In order to accurately identify the sensitive information in the intangible cultural heritage communication network and realize the reasonable protection of intangible cultural heritage data, a method for protecting t...
详细信息
the proceedings contain 152 papers. the special focus in this conference is on machinelearning for Cyber Security. the topics include: machinelearning Based Security Situation Awareness Method for Network Data Trans...
ISBN:
(纸本)9783031200984
the proceedings contain 152 papers. the special focus in this conference is on machinelearning for Cyber Security. the topics include: machinelearning Based Security Situation Awareness Method for Network Data Transmission Process;multi-objective Hydrologic Cycle Optimization for Integrated Container Terminal Scheduling Problem;High Voltage Power Communication Network Security Early Warning and Monitoring System Based on HMAC Algorithm;Large Scale Network Intrusion Detection Model Based on FS Feature Selection;Privacy Preserving CSI Fingerprint Device-Free Localization;Research on Intelligent Detection Method of Automotive Network Data Security Based on FlexRay/CAN Gateway;adversarial Attack and Defense on Natural Language Processing in Deep learning: A Survey and Perspective;a Novel Security Scheme for Mobile Healthcare in Digital Twin;Construction of Security Risk Prediction Model for Wireless Transmission of Multi Axis NC Machining Data;spiking Neural Networks Subject to Adversarial Attacks in Spiking Domain;Diverse Web APIs Recommendation with Privacy-preservation for Mashup Development;network Security Evaluation Method of College Freshmen Career Counseling Service Based on machinelearning;FedTD: Efficiently Share Telemedicine Data with Federated Distillation learning;increase Channel Attention Based on Unet++ Architecture for Medical Images;distributed Power Load Missing Value Forecasting with Privacy Protection;A Novel Blockchain-MEC-Based Near-Domain Medical Resource Sharing Model;differentially Private Generative Model with Ratio-Based Gradient Clipping;differential Privacy Protection Algorithm for Data Clustering Center;improved Kmeans Algorithm Based on Privacy Protection;symmetry Structured Analysis Sparse Coding for Key Frame Extraction;data Reconstruction from Gradient Updates in Federated learning;natural Backdoor Attacks on Speech Recognition Models;boarding Pass Positioning with Jointly Multi-channel Segmentation and Perspective Transformation Co
the proceedings contain 40 papers. the topics discussed include: analyzing the use of auditory filter models for making interpretable convolutional neural networks for speaker identification;leveraging model driven te...
ISBN:
(纸本)9798350338195
the proceedings contain 40 papers. the topics discussed include: analyzing the use of auditory filter models for making interpretable convolutional neural networks for speaker identification;leveraging model driven techniques for designing Web-GIS systems;multi-objective optimization for neural network structure;a data-centric approach for improving adversarial training through the lens of out-of-distribution detection;biological signals for diagnosing sleep stages using machinelearning models;SBSN: harvesting stable body sensor node by providing an energy efficient adaptive sampling method;an approximate method for spatial task allocation in partially observable environments;an ontology-based approach to facilitate semantic interoperability of context-aware systems;predicting the load capacity of 4G cellular networks with deep learning;predicting users' demographic features based on searched queries and installed apps and games;fashion compatibility learning via triplet-Swin transformer;and a question summarization method based-on deep learning in Persian language.
暂无评论